ORGANIC LETTERS
A Convenient Synthesis of 4-Alkyl-5-aminoisoxazoles
2006 Vol. 8, No. 17 3679-3680
Matthew P. Bourbeau* and James T. Rider Department of Medicinal Chemistry, Amgen Inc., 1 Amgen Center DriVe, Thousand Oaks, California 91303
[email protected] Received May 23, 2006
ABSTRACT
A series of 4-alkyl-5-aminoisoxazoles have been synthesized in high yield by nucleophilic addition of lithiated alkyl nitriles to (r)-chlorooximes. The scope and limitations of this reaction were examined by varying the nature of the nitrile and chloride oxime.
Isoxazoles have attracted a great deal of interest as synthetic targets.1 In particular, the synthesis of 5-aminoisoxazoles has been achieved by a variety of different methods. The most common method for the synthesis of 5-aminoisoxazoles, described by Nishawiki, involves the condensation of esters with nitrile anions, followed by cyclization of the resulting β-ketonitrile with hydroxylamine under basic conditions at reflux to afford the desired isoxazole.2 Additional methods for the synthesis of 5-aminoisoxazoles include the addition of stabilized nitrile anions to (R)-chlorooximes (resulting in 5-aminoisoxazoles with electron withdrawing groups at the C-4 position), condensation of R-bromo ketoximines with cyanide, and rearrangements of 5-alkyl-4-cyanoisoxazoles upon treatment with LiAlH4.3-5 During a recent synthetic investigation, we examined the synthesis of 5-aminoisoxazoles with an aryl substituent at the 3 position. We were also interested in incorporating alkyl groups at the C-4 position. Initially, we planned to follow Nishawiki’s procedure for the addition of nitriles to esters, followed by cyclization with hydroxylamine.2 However, there was a specific case that we encountered where the addition of lithiated acetonitrile to an aryl ester did not provide the desired product. To address this problem, we were attracted to the work of Beccalli, who showed that the sodium anions of alkyl nitriles bearing additional electron withdrawing (1) Baraldi, P. G.; Barco, A.; Bennetti, S.; Pollini, G. P.; Simoni, D. Synthesis 1987, 857-869. (2) Nishiwaki, T.; Saito, T. J. Chem. Soc. (C) 1971, 3021-3026 10.1021/ol061260+ CCC: $33.50 Published on Web 07/20/2006
© 2006 American Chemical Society
groups R to the nitrile add to (R)-chlorooximes at elevated temperatures, affording the desired isoxazole.3 To the best of our knowledge there has been no report in the literature regarding the addition of nitrile anions lacking additional electron withdrawing groups to (R)-chlorooximes. We reasoned that (R)-chlorooximes might be sufficiently electrophilic to undergo addition with lithiated nitriles at low temperature. Gratifyingly, we found that when benzaldehyde (R)-chlorooxime6 was treated with an excess of lithiated acetonitrile, the desired 5-aminoisoxazole was isolated in reasonable yield (eq 1).
Encouraged by this result, we examined this addition with a variety of substituted nitriles (Table 1). The reaction was tolerant of additional alkyl functionality, thus allowing for (3) Beccalli, E. M.; Manfredi, A.; Marchesini, A. J. Org. Chem. 1985, 13, 2372-2375. (4) Kong, W. C.; Kim, K.; Park, Y. J. Heterocycles 2001, 55, 75-89. (5) Alberola, A.; Gonzalez, A. M.; Laguna, M. A.; Pulido, F. J. J. Org. Chem. 1984, 49, 3423-3424. (6) Lam, P. Y. S.; Adams, J. J.; Clark, C. G.; Calhoun, W. J.; Luettgen, J. M.; Knabb, R. M.; Wexler, R. R. Bioog. Med. Chem. Lett. 2003, 13, 1795-1799.
Table 1. The Condensation of Lithiated Nitriles with Benzaldehyde (R)-Chlorooximea
a
entry
R
yield (%)
1a 1b 1c 1d 1e 1f
H Me i-Pr cyclopropyl Ph Bn
51 89 64 78 86 91
All reactions unoptimized.
the synthesis of C4-alkylisoxazoles. Increasing the size of the alkyl group from methyl (entry 1b) to isopropyl (entry 1c) and cyclopropyl (entry 1d) showed little effect on the reaction yield. Additionally, 2-phenylacetonitrile (entry 1e) and 3-phenylpropionitrile (entry 1f) were also good substrates for this condensation. Having examined the tolerability of nitrile functionality in the reaction, we next examined the reaction of lithiated propionitrile with a variety of different (R)-chlorooximes (Table 2). Substitution of the phenyl ring of benzaldehyde
with a small amount (∼5%) of 3-(2-chlorothiophen-3-yl)4-methylisoxazol-5-amine as a side product.7 Picolinaldehyde (R)-chlorooxime (entry 2h) similarly reacted in lower yield. In considering a mechanistic rational for this process, two potential pathways have been proposed. The first invokes a stepwise process starting with addition of the lithiated nitrile to a nitrile oxide generated by deprotonation of the (R)chlorooxime, followed by cyclization to form the 5-aminoisoxazole. Alternatively, the reaction may proceed through a concerted pathway, analogous to reactions between (R)chlorooximes and alkynes.1 In summary, the reaction of lithiated nitriles with (R)chlorooximes has been found to be useful for synthesizing a variety of 5-aminoisoxazoles, particularly ones with alkyl functionality at the 4 position. As (R)-chlorooximes have been shown to also be useful substrates for the synthesis of 5-alkylisoxazoles, this reaction would potentially be useful to chemists seeking to generate libraries of both 5-alkyl- and 5-aminoisoxazoles from the same set of starting materials (Figure 1). Additionally, as both Nishawiki and Becalli’s
Figure 1. The use of an (R)-chlorooxime as a common intermediate for the synthesis of 5-aminoisoxazoles and 5-alkylisoxazoles. Table 2. The Addition of Lithiated Propionitrile to a Variety of (R)-Chlorooximesa
entry
R
yield (%)
2a 2b 2c 2d 2e 2f 2g 2h
4-methoxyphenyl 4-chlorophenyl 3-chlorophenyl 2-chlorophenyl cyclohexyl n-propyl 3-thienyl 2-pyridyl
77 73 81 69 53 74 34b 49
a All reactions unoptimized. b A small amount (
